中文摘要：

不同来源的矢量数据集合并时，必须以相匹配的同名点为基础进行数据集校准，同名点匹配结果的可靠性直接影响数据集校准的正确性，目前还没有一个有效的方法判别同名点是否正确地被匹配。为确保输出结果确实源于各数据集中的同名点，必须对同名点坐标输出结果的可靠性进行统计检验。该文基于叠置方法对来自不同矢量数据集中的同名点特征匹配结果进行检验，以确保点特征匹配结果的可靠性。实践证明，通过显著性检验，可确保数据集校准结果的正确性，探测错误匹配点的位置，能较好地控制Rubber-Sheeting变换处理。

英文摘要：

When the vector data sets from different geospatlal resources are merged using the teature point--based matctamg, me data conversion and correlation calculation are imperative and essential. In order to ensure the high--quallty of the confused and merged data set, the post--statistic verification of the resulted data should be investigated. This paper presents a statistic method called overlay--based to verify the different vector data sources when using the feature point--based matching in map conflatlon. With the proposed method, a boundary llne from the four maps with different scales, 1：500,1：1 000,1：2 000, and 1：5 000, is tested. The results show that the proposed method can correctly detect the mismatching points, and improve the accuracy of the Rubber--Sheeting method, thus, the reliability of map alignment can be largely improved.